I'm currently training the next model I think this next one looks much harder to tell.